We propose to conduct Phase I clinical trials of new anticancer agents or combinations of anticancer agents for the purposes of characterizing drug toxicity, determining the maximum tolerated dose, characterizing the pharmacokinetics and relating the clinical endpoints to the derived pharmacokinetic parameters and/or relevant biologic endpoints. While the endpoints are dependent on the therapeutic intent of the specific drug under study, for cytotoxic agents in general this will be the maximum tolerated dose. The endpoints for the evaluation of biologic agents alone or in combination with other agents will be those judged to be induced in model systems which serve as the rationale for advancement to phase I clinical trial. The specific aims of these studies are to determine as efficiently as is compatible with patient safety the appropriate dose of new anticancer agents selected by the National Cancer Institute for evaluation of therapeutic activity in subsequent Phase II trials; to identify pharmacokinetic parameters which may predict toxicity; to develop limited sampling models which may be used in phase II trials; and to evaluate biologic endpoints which may be used as surrogate markers of drug effect. Studies of drug combinations will be designed based on preclinical data identifying the appropriate schedule and exposure duration of the agents. Studies of novel modulations of drug effect will be planned to monitor and evaluate the intended modulation. Each new agent will be evaluated using a standard modified Fibonacci dose escalating scheme or using the modified continual reassessment method of dose escalation in a phase I trial. The studies will be designed to incorporate the appropriate pharmacokinetic analysis, drug analysis, and pharmacodynamic assessment. Pharmacokinetic/ Pharmacodynamic modelling of the data and development of limited sampling models will be performed in those studies in which it is appropriate.